Cyberbullying and Social Anxiety: a Latent Class Analysis Among Spanish Adolescents
Total Page:16
File Type:pdf, Size:1020Kb
International Journal of Environmental Research and Public Health Article Cyberbullying and Social Anxiety: A Latent Class Analysis among Spanish Adolescents María C. Martínez-Monteagudo 1 , Beatriz Delgado 1,* ,Cándido J. Inglés 2 and Raquel Escortell 3 1 Department of Developmental Psychology and Didactic, University of Alicante, 03690 Alicante, Spain; [email protected] 2 Department of Health Psychology, Miguel Hernandez University of Elche, 03202 Alicante, Spain; [email protected] 3 Faculty of Education, International University of La Rioja, 26006 Logrono, Spain; [email protected] * Correspondence: [email protected]; Tel.: +34-965-903-495 Received: 29 November 2019; Accepted: 6 January 2020; Published: 8 January 2020 Abstract: Cyberbullying is a common social maladjustment that has negative repercussions on the wellbeing and development of adolescents, but numerous questions remain as to the relationship between cyberbullying and social anxiety in adolescence. This study analyzes cyberbullying profiles (screening of harassment among peers) and assesses whether these profiles vary with respect to the level of social anxiety (social anxiety scale for adolescents). The sample consisted of 1412 Spanish secondary education students aged 12 to 18 (M = 14.36, SD = 1.65). Latent class analysis and ANOVA were performed. Analyses revealed three profiles: high cyberbullying (high victimization, aggression, and aggression-victimization), low cyberbullying (moderate victimization, aggression, and aggression-victimization), and non-cyberbullying. The cyberbullying patterns varied significantly for all social anxiety subscales. Students with the high cyberbullying profile (bully–victims) presented high scores on social avoidance and distress in social situations in general with peers, whereas these students presented lower levels of fear of negative evaluation and distress and social avoidance in new situations as compared to the low cyberbullying (rarely victim/bully) and non-involved student profiles. Implications for psychologists and educational counselors and cyberbullying preventive interventions are discussed. Keywords: cyberbullying; victimization; aggression; social anxiety; adolescence 1. Introduction Over the past decade, a major increase has been seen in news of bullying carried out by school-aged children using the new information and communication technologies (ICT). The ICT and social networks have become indispensable communication tools, especially for youth. This widespread use has offered many advantages; however, it has also led to some new violent behaviors resulting from the inappropriate use of these technologies. So, some students having a great domain of the ICT have taken advantage of these new virtual scenarios to engage in aggressive behavior towards their peers (such as insults, humiliation, coercion, the publication of confidential information, threats, denigration, violation of privacy, social exclusion, the spreading of rumors, identity theft, the dissemination of physical assaults, etc.). This phenomenon, known as cyberbullying, is defined as “a type of aggressive and intentional behavior that repeats frequently over time through the individual or group use of electronic devices with a victim that is unable to easily defend him/herself” ([1], p. 376). The prevalence of cyberbullying has varied considerably in the studies that have been carried out until now. International reviews have reported mean prevalences ranging from 4% to 36% for Int. J. Environ. Res. Public Health 2020, 17, 406; doi:10.3390/ijerph17020406 www.mdpi.com/journal/ijerph Int. J. Environ. Res. Public Health 2020, 17, 406 2 of 13 cybervictimization and 16 to 18% for cyberaggression [2,3]. In a recent meta-analysis, Modecki, Minchin, Harbaugh, Guerra, and Runions (2014) found variations in prevalence ranging from 5 to 32% for cyberaggressors (mean of 16%) and between 2 and 56% for victims of cyberbullying (mean of 15%) [4]. These large variations may be due to differing conceptualizations of cyberbullying, the cut-off point criteria used to establish the frequency, the time framework established (an incident taking place during the past two months, last year, at any time, etc.), the type of methodology used, sample age range, etc. 1.1. Roles in Cyberbullying The scientific literature has established three main roles with regard to this issue: victims, aggressors, and non-involved [5,6], with this being the most parsimonious classification. It is possible for aggressors to have previously been victims, with these students becoming aggressors in an attempt to earn a reputation of being strong and capable of defending themselves, and thus, the aggressor/victim role is created [6–9]. Studies have analyzed student roles in cyberbullying through the creation of cut-off scores that are based on statistical distributions that permit the assignment of the participants to one of these roles [5,10,11]. So, modification of the cut-off points alters the number of students belonging to a specific group, such that the stricter the established cut-off point, the lower the proportion of student aggressors [10], suggesting that these cut-off points may be relatively arbitrary. This problem may be overcome using person-centered analytical approaches, such as cluster analysis and latent class analysis (LCA). With these analyses, student groups are generated based on specific indicators, permitting the creation of distinct groups based on the students’ real participation, with members of the same group having similar experiences that are distinct from those of other groups to which they do not belong. So, Aoyama, Bernard-Brak, and Talbert (2011), using cluster analysis with a sample of 133 US adolescents, identified four groups of roles involved in cyberbullying. The majority of the sample belonged to the “least involved” group (51.1%), 12.8% were “highly involved as bully and victim”, 10.5% were “more bully than victim”, and 9.8% were “more victim than bully” [12]. Along these lines, Schultze-Krumbholz et al. (2015), using LCA in an extensive sample of 6260 youth from six European countries, found that the majority of the sample belonged to the “non-involved” group (70.1%), while the “bully/victim” group was made up of 26.1% of the students and, a last group, the so-called “perpetrator with mild victimization” group, consisted of 4% of the selected sample [13]. Barboza (2015) used LCA to identify four categories: “highly victimized by both bullying and cyber bullying behaviors” (3.1%); “victims of relational bullying, verbal bullying, and cyber bullying” (11.6%); “victims of relational bullying, verbal bullying, and physical bullying but were not cyber bullied” (8%); and “non-victims” (77.3%) [14]. Hollá (2016), found three groups of students using LCA in a sample of 1619 Slovakian children and adolescents aged 11 to 18. Here, 52.9% of the students belonged to the “uninvolved” group while 42.7% were part of the “victims” group and 4.4% belonged to the “victims–aggressors” group [15]. In a more recent study, Betts, Gkimitzoudis, Spenser, and Baguley (2017), using a sample of 440 British students aged 16 to 19, identified four student profiles using a cluster analysis technique: “not involved” (33%), “rarely victim and bully” (40%), “typically victim” (26%), and “retaliator” (1%) [16]. In a subsequent study, Schultze-Krumbholz, Hess, Pfetsch, and Scheithauer (2018) used LCA on a sample of 849 German students (11 to 17 years of age), determining five groups: “prosocial defenders”, “communicating outsiders”, “aggressive defenders”, “bully–victims”, and “assistants” [17]. A summary of studies on cyberbullying with a person-centered analytical approach are presented in Table1. So, past empirical research supports the presence of distinct profiles in relation to cyberbullying. However, the results of these studies differ, most likely due to the distinct conceptualizations of cyberbullying, the type of methodology used, or the frequency considered necessary to consider the behavior “cyberbullying”. These inconclusive results support the need to continue analyzing the roles Int. J. Environ. Res. Public Health 2020, 17, 406 3 of 13 in cyberbullying to offer greater clarification about its functioning and to develop actions aimed at the effective prevention of the cyberbullying. Table 1. Summary of the studies reviewed. Source Country Subjects Method Classes 51.1% least involved 12.8% 133 high school highly bully and Aoyama et al., 2011 USA students Cluster analysis victim 10.5% more (Mage = 15.7) bully than victim 9.8% more victim than bully 77.3% non-victims 11.6% victims of relational and verbal bullying and cyberbullying 8% 5589 students victims of Barboza, 2015 USA LCA (aged 12–18) relational, verbal and physical bullying 3.1% highly victimized by both bullying and cyberbullying 33% not involved 40% rarely victim 440 students Betts et al., 2017 United Kingdom Cluster analysis and bully 26% (aged 16–19) typically victim 1% retaliator 52.9% uninvolved 1619 students Hollá, 2016 Slovakia LCA 42.7% victims 4.4% (aged 11–18) victims–aggressors 70.1% Poland, Spain, Italy, non-involved 26.1% Schultze-Krumbholz 6260 students United Kingdom, LCA bully/victim 4% et al., 2015 (aged 11–23) Germany, Greece perpetrator with mild victimization 52.2% prosocial defenders 28.4% communicating Schultze-Krumbholz 849 students outsiders 9.5% Germany LCA et al., 2018 (aged 11–17) aggressive defenders 7.1% bully–victims 2.8% assistants 1.2. Cyberbullying Roles and Social Anxiey Numerous studies